Summary of Work: A major focus of this project is to discover the role of the Amyloid Precursor Protein (APP) in the etiology and pathology of Alzheimer's Disease (AD). The normal physiological role of this protein is also under investigation. APP is important to study since the processing of APP and the effect of APP mutations and Presenilin mutations on APP processing bear directly on the increased production and extracellular deposition of A-beta peptides in AD. The alpha- and beta-secretase processing of APP also generates two large N-terminal secreted forms of the protein, which may have neurotrophic or neurotoxic properties, respectively. Brains of AD patients exhibit decreased synaptic connectivity and selective and massive neuronal cell loss. We are interested in examining the mechanisms involved in this cell death. Rare mutations in APP lead to an early onset, autosomal dominant form of AD (FAD). Previously, we showed that over-expression of FAD mutant forms of APP by either stably transfecting PC12 cells or by adenovirus-mediated gene transfer of primary cortical neurons led to increased apoptotic cell death over several days. More recently, we have been studying the toxic effects of several A-beta peptides on human neuroblastoma cells and the signaling pathways that are activated upon A-beta peptide treatment. Low concentrations of A-beta 1-42 killed SH-SY5Y and IMR-32 cells by apoptosis as measured by an ELISA. A-beta 1-40 was much less potent. A-beta 17-42, derived from APP by sequential alpha- and gamma-secretase cutting of APP, dose-dependently killed these cells apoptotically. A-beta 17-42 previously was thought to be a product of non-amyloidogenic APP processing and not a toxic peptide. Recent evidence shows that this peptide accumulates in the plaques of AD brains although its importance in AD pathology is unclear at this time. A-beta17-42 activated caspase-8 and caspase-3, and induced PARP cleavage, proteins important in a cascade of events leading to apoptotic cell death. Selective caspase-8 and caspase-3 inhibitors completely blocked A-beta 17-42 induced neuronal death. A-beta 17-42 activated c-jun N-terminal kinase (JNK) approximately two fold. Over-expression of a dominant-interfering SEK1 construct (a protein kinase that phosphorylates and activates JNK) protected against A-beta 17-42 -induced neuronal death by 50-70 %. The results demonstrate that A-beta 17-42 peptide induced neuronal apoptosis via a Fas-like/caspase-8 activation pathway. The results also suggest that p3 peptide may be an additional toxic peptide derived from APP proteolysis and may have a role in the neuronal cell loss characteristic of AD. We also analyzed the early signaling mechanisms of A-beta toxicity using human neuronal SH-SY5Y cells. We have focused on mitogen-activated protein kinases and the PI-3 kinase/Akt cascades. A-beta 1-42 treatment, which resulted in a dose-dependent cell death, weakly activated Akt and ERK, but had no effect on p38 kinase. However, this activation of ERK and Akt by A-beta 1-42 apparently did not play a role in A-beta toxicity since specific inhibitors of these kinase pathways, U0126 and wortmannin respectively, had no influence on A-beta-induced neuronal death. However A-beta 1-42-induced JNK activation by about two fold and seemed to play a critical role in A-beta-induced neuronal death since the dominant-negative construct SEK-1 blocked JNK activation and protected against cell death. Insulin-like growth factor-1 (IGF-1) dose-dependently protected cells from A-beta toxicity by strongly activating ERK and Akt and blocking A-beta-induced JNK activation. A specific Go/Gi inhibitor, pertussis toxin, (PT) also protected against A-beta toxicity by blocking A-beta-induced JNK activation. These results suggested that A-beta peptides in part could activate PT-sensitive-G-proteins leading to JNK activation. This may be an important and early event of A-beta toxicity and an early signaling pathway underlying neuronal loss in AD pathology. Finally, we have been examining the effects of a shortened, secreted form of APP derived from the initial beta-secretase processing of APP. This protein is called secreted APP beta (sAPPb). We generated cell lines over-expressing authentic sAPPa and sAPPb and which secrete these proteins into the surrounding media. We found that conditioned media containing sAPPb, when added to either NGF differentiated PC12 cells or primary cortical neurons, led to an apoptotic cell death, while conditioned media containing sAPPa had no such effect. An antibody specific to sAPPb prevented the cell death and a slightly truncated, highly purified form of sAPPb also caused cell death. These latter results strongly suggested that sAPPb itself is an additional toxic protein derived from beta-secretase APP processing. Overall these results provide a rationale for targeting particular elements of apoptotic pathways as well as APP processing for therapeutic intervention in AD.